EP3968606B1 - Method for deploying services and scheduling apparatus - Google Patents

Method for deploying services and scheduling apparatus Download PDF

Info

Publication number
EP3968606B1
EP3968606B1 EP20834405.1A EP20834405A EP3968606B1 EP 3968606 B1 EP3968606 B1 EP 3968606B1 EP 20834405 A EP20834405 A EP 20834405A EP 3968606 B1 EP3968606 B1 EP 3968606B1
Authority
EP
European Patent Office
Prior art keywords
qos
execution
path
service
preset
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP20834405.1A
Other languages
German (de)
English (en)
French (fr)
Other versions
EP3968606A1 (en
EP3968606A4 (en
EP3968606C0 (en
Inventor
Lei Xiao
Bo Jin
Si Shen
Youqiang LI
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Cloud Computing Technologies Co Ltd
Original Assignee
Huawei Cloud Computing Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Cloud Computing Technologies Co Ltd filed Critical Huawei Cloud Computing Technologies Co Ltd
Publication of EP3968606A1 publication Critical patent/EP3968606A1/en
Publication of EP3968606A4 publication Critical patent/EP3968606A4/en
Application granted granted Critical
Publication of EP3968606B1 publication Critical patent/EP3968606B1/en
Publication of EP3968606C0 publication Critical patent/EP3968606C0/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/61—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources taking into account QoS or priority requirements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/51—Discovery or management thereof, e.g. service location protocol [SLP] or web services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
    • H04L41/5054—Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/80—Actions related to the user profile or the type of traffic
    • H04L47/805—QOS or priority aware
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/34—Network arrangements or protocols for supporting network services or applications involving the movement of software or configuration parameters 
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services

Definitions

  • This application relates to the field of cloud computing, and in particular, to a service deployment method, a scheduling apparatus and a computer-readable storage medium.
  • a scheduling apparatus is usually deployed on a cloud platform, and is configured to select, based on a requirement of a to-be-deployed service, an execution path for deploying the to-be-deployed service. Specifically, the scheduling apparatus selects, based on a type of the to-be-deployed service, at least one execution device that meets the requirement of the to-be-deployed service and that is configured to execute the to-be-deployed service, to determine an execution path including the at least one selected execution device, and deploy the to-be-deployed service on the determined execution path.
  • the scheduling apparatus selects, based on quality of service (QoS) of each execution device, an execution device that meets the requirement of the to-be-deployed service.
  • QoS quality of service
  • the scheduling apparatus focuses only on QoS of one or several execution devices. This is because different to-be-deployed services have different QoS requirements for execution devices. For example, a to-be-deployed communications service has a relatively high QoS requirement for an execution device used for forwarding, and a relatively low QoS requirement for an execution device used for computing. Conversely, a to-be-deployed financial service has a relatively low QoS requirement for an execution device used for forwarding, and a relatively high QoS requirement for an execution device used for computing.
  • the execution path is selected based on the QoS of the one or several execution devices. Consequently, the execution path selected by the scheduling apparatus can only roughly meet the requirement of the to-be-deployed service, but may not fully meet the requirement of the to-be-deployed service.
  • US 2008/0181118 A1 discloses that service nodes in a service overlay network are selected to provide a service path for a requested service by determining a reduced search space of service nodes in the service overlay network that are operable to provide service components for the requested service. From the reduced search space of service nodes, service nodes are selected to form a service path that satisfies all quality of service constraints for the service.
  • This application provides a service deployment method, a scheduling apparatus and a computer-readable storage medium, as defined in the appended set of claims, to calculate QoS of execution paths, and select, based on the QoS of the execution paths, an execution path for deploying a to-be-deployed service. This makes execution path selection more accurate and avoids a case in which the selected execution path cannot meet a requirement of the to-be-deployed service.
  • the preset QoS of the to-be-deployed service is preset in the scheduling apparatus by a user, and after the user sets the preset QoS, the scheduling apparatus obtains the preset QoS of the to-be-deployed service.
  • This application first provides a solution in which the scheduling apparatus determines, based on QoS of execution devices on an execution path, a capability of the execution path to provide a service, obtains the QoS of execution path, and selects, based on the QoS of the execution path, an execution path for deploying the to-be-deployed service, to determine, based on the QoS of the execution devices on the execution path, the capability of the execution path to provide a service.
  • the capability namely, the QoS of the execution path
  • the scheduling apparatus further obtains the QoS of the transmission link between the execution devices, and determines the QoS of the execution path based on the QoS of the execution devices and the QoS of the transmission link, so that the calculated QoS of the entire execution path is more accurate.
  • This application specifically describe how the scheduling apparatus obtains the QoS of the execution device (that is, the scheduling apparatus obtains the QoS of the execution device by sending the corresponding QoS calculation formula to each execution device). This is practical.
  • the scheduling apparatus further determines the importance degree of each execution device on the execution path based on the weight of the QoS of the execution device. This is more targeted.
  • This application further illustrate that when a relationship between the QoS of each of the N execution paths and the preset QoS does not meet the preset rule, the scheduling apparatus performs resource adjustment, to deploy the to-be-deployed service on the adjusted execution path. This ensures reliability and stability of the to-be-deployed service.
  • the scheduling apparatus not only compares QoS of each execution path in the current periodicity with the preset QoS, but also compares QoS of each execution path in the next periodicity with the preset QoS.
  • the scheduling apparatus deploys the to-be-deployed service on the execution path only when a relationship between the QoS of the execution path in the current periodicity and the preset QoS and a relationship between the QoS of the execution path in the next periodicity and the preset QoS meet a preset relationship. This further ensures stability of the to-be-deployed service.
  • Cloud computing is a new business service model and computing model, and provides, through the internet, various computing and storage services for a user who accesses the cloud computing. These services do not rely on a local computer, but rely on at least one cloud server.
  • the cloud server provides computing, storage, and network resources and provides services for the user based on the resources provided by the cloud server.
  • resources on the cloud server may be allocated to the to-be-deployed service based on a requirement of the to-be-deployed service. That is, based on the on-demand allocation feature of the cloud server, the resources of the cloud server may be allocated based on task requirements.
  • One manner is resource allocation based on different priority policies. That is, a to-be-deployed service with a high priority obtains more resources, and a to-be-deployed service with a low priority obtains fewer resources.
  • sufficient resources are allocated to the to-be-deployed service with a high priority for execution. This can ensure that the user has relatively high satisfaction with an execution result of the to-be-deployed service, but resources are often excessive and a waste of resources is often caused.
  • the to-be-deployed service with a low priority is usually in a starvation state due to insufficient resources, and this cannot meet basic user experience.
  • the other manner is QoS-based resource allocation.
  • the resources are allocated based on QoS required by a specific to-be-deployed service.
  • QoS indicates a capability of an execution device or a service to provide a service for a specific to-be-deployed service, and the QoS is a QoS value.
  • an execution path is selected based on QoS of one or several execution devices. Consequently, the execution path selected by a scheduling apparatus can only roughly meet a requirement of the to-be-deployed service, but may not fully meet the requirement of the to-be-deployed service.
  • a scheduling apparatus is configured to allocate resources of a cloud server based on a user request (for example, an entered to-be-deployed service) entered by a user, and is specifically responsible for processing a task entered by the user, saving status data, scheduling resources, and the like.
  • the scheduling apparatus centrally schedules, based on the user request, resources included in execution devices that are involved in the user request on the cloud server.
  • the scheduling apparatus may schedule, based on specifications of resources required by the to-be-deployed service, resources included in an execution device 1, an execution device 2, ..., and an execution device n (it is assumed that the execution devices involved in the to-be-deployed service are the execution device 1, the execution device 2, ..., and the execution device n). It should be noted that a minimum functional unit on the cloud server may be referred to as an execution device, and specifications of resources included in the execution device are not lower than the specifications of the resources required by the to-be-deployed service.
  • the execution device represents a candidate device for deploying the to-be-deployed service
  • the execution path represents a candidate path for deploying the to-be-deployed service.
  • the execution device may include a computing device, a storage device, a network device, and the like.
  • the computing device may include a physical machine, a virtual machine, a container, and the like.
  • the scheduling apparatus and a functional apparatus that executes a specific service may be deployed on a same cloud server. That is, corresponding software/hardware is installed on the cloud server to implement resource scheduling.
  • the scheduling apparatus may be deployed on a plurality of different cloud servers. For example, a plurality of virtual machines jointly serve as the scheduling apparatus to implement resource scheduling.
  • the scheduling apparatus may be independently deployed (the scheduling apparatus shown in FIG. 1 is independently deployed). This is not specifically limited herein.
  • the execution path includes execution devices through which data related to a service passes when the service is completed and transmission links between the execution devices.
  • FIG. 2 is used as an example for description. If a service B is a storage service, and needs to be executed, data related to the service B first arrives at a virtual private cloud (VPC), and is forwarded in the VPC to a virtual tunnel endpoint (VTEP) 1 of the VPC.
  • VPC virtual private cloud
  • VTEP 1 is connected to a VTEP 2
  • the VTEP 2 may be a physical switch.
  • An example of a physical switch such as a top of rack (ToR) is drawn in FIG. 2 .
  • the data arrives at the ToR through a transmission link 1 formed between the VTEP 1 and the VTEP 2, then arrives at a bare metal host 1 through a transmission link 2, and finally arrives at a final storage area network (SAN) through a transmission link 3.
  • execution of the service B is completed.
  • the VPC, the ToR, the bare metal host 1, the SAN, the transmission link 1 between the VPC and the ToR, the transmission link 2 between the ToR and the bare metal host 1, and the transmission link 3 between the bare metal host 1 and the SAN form an execution path (which may be referred to as a path b1) of the service B.
  • the VPC, the ToR, the bare metal host 1, and the SAN are execution devices on the execution path, and the execution path of the service includes at least one execution device.
  • a scheduling apparatus may determine the execution path of the to-be-deployed service in a plurality of manners. This is not specifically limited herein. The following illustrates one of the determining manners:
  • the scheduling apparatus receives a user request, where the user request carries a service identifier of the to-be-deployed service.
  • the scheduling apparatus obtains the service identifier, and may determine, based on the service identifier, a type (there is a mapping relationship between the service identifier and the type of resources required by the to-be-deployed service) of the resources required by the to-be-deployed service that needs to be executed by the user.
  • the scheduling apparatus further needs to obtain specifications of the resources required by the to-be-deployed service, and then determines the execution path based on the type and the specifications of the resources required by the to-be-deployed service.
  • a service B is used as an example.
  • the service B is a storage service. It is assumed that the service B requires 2 GB storage resources (that is, specifications of resources required by the service B are 2 GB).
  • the scheduling apparatus can obtain an identifier (it is assumed that the service identifier of the service B is "1122") of the service B at the same time.
  • the scheduling apparatus can determine, based on the service identifier " 1122", that a type of the resources required by the service B is storage resources, and the specifications of the resources required by the service B are 2 GB.
  • the scheduling apparatus determines execution devices (for example, execution devices such as a VPC, ToR, a bare metal host 1, and a SAN that can provide 2 GB storage resources) that are on the cloud server and that can provide 2 GB storage resources, so that the scheduling apparatus determines a path (namely, the path b1 in FIG.
  • the execution devices such as the VPC, the ToR, the bare metal host 1, the SAN and a transmission link 1, a transmission link 2, and a transmission link 3 between the VPC, the ToR, the bare metal host 1, and the SAN, as a path (which may also be referred to as an execution path) on which the to-be-deployed service can be executed.
  • the scheduling apparatus further determines whether to use the execution path as an execution path for deploying the to-be-deployed service.
  • the scheduling apparatus After determining the execution path, the scheduling apparatus further obtains the QoS of each execution device on the execution path.
  • the QoS of the execution device is determined based on specifications of resources included in the execution device. A quantity of resources that the execution device can provide for the to-be-deployed service is referred to as the specifications of the resources included in the execution device.
  • the scheduling apparatus may obtain the QoS of each execution device in a plurality of manners. This is not specifically limited herein. The following illustrates several manners of obtaining the QoS of each execution device on the execution path:
  • the scheduling apparatus may send a QoS calculation formula that corresponds to each execution device (that is, one execution device corresponds to one QoS calculation formula) to a corresponding execution device, and each execution device calculates QoS of the execution device according to the received QoS calculation formula.
  • QoS of the physical machine may be calculated according to a calculation formula of central processing unit (CPU) usage, memory usage, a bandwidth, a latency, and the like of a host.
  • CPU central processing unit
  • each execution device sends the QoS of the execution device that is calculated according to the QoS calculation formula to the scheduling apparatus.
  • the QoS calculation formula corresponding to the execution device may be preset in the scheduling apparatus.
  • the scheduling apparatus sends the QoS calculation formula of each execution device to the corresponding execution device.
  • the QoS calculation formula may also be stored in a device A (including a virtual device, a physical device, and the like) connected to the scheduling apparatus.
  • the scheduling apparatus may send a request instruction to the device A that is connected to the scheduling apparatus, where the request instruction is used to trigger the device A to send the QoS calculation formula corresponding to each execution device on the execution path to the scheduling apparatus. Then the scheduling apparatus sends each QoS calculation formula to a corresponding execution device.
  • the QoS calculation formula may alternatively be preset in each execution device. After determining the execution path, the scheduling apparatus sends an instruction to each execution device on the execution path. The instruction is used to trigger each execution device to calculate QoS of the execution device according to the preset QoS calculation formula of the execution device. In this embodiment of this application, how the execution device obtains the QoS calculation formula is not limited.
  • the scheduling apparatus may have a mapping relationship library.
  • the mapping relationship library stores the mapping relationship between the specifications of the resources included in each execution device and the QoS of the corresponding execution device.
  • the scheduling apparatus may obtain specifications of resources included in each execution device at a current moment, and find, in the mapping relationship library, QoS of the execution device at the current moment based on the mapping relationship between the specifications of the resources included in the execution device and the QoS of the execution device.
  • the following uses an example for description.
  • execution devices on the path c are a device a, a device b, and a device c. In this case, the scheduling apparatus needs to obtain QoS of the device a, the device b, and the device c.
  • the scheduling apparatus first obtains specifications of resources included in the device a, the device b, and the device c at the current moment (it is assumed that the specifications of the resources included in the device a, the device b, and the device c at the current moment are x1, y2, and z3 respectively), and finds, in the mapping relationship library (Table 1 shows a mapping relationship library), the corresponding QoS (it is assumed that the QoS of the device a is a-QoS-1) of the device a when the specifications of the resources included in the device a are x1, the corresponding QoS (it is assumed the QoS of the device b is b-QoS-2) of the device b when the specifications of the resources included in the device b are y2, and the corresponding QoS (it is assumed the QoS of the device c is c-QoS-3) of the device c when the specifications of the resources included in the device c are z3.
  • Table 1 shows a mapping relationship library
  • the scheduling apparatus obtains the QoS of the device a, the device b, and the device c at the current moment.
  • Table 1 Mapping relationship library Device a x1 ⁇ a-QoS-1 x2 ⁇ a-QoS-2 x3 ⁇ a-QoS-3 ...
  • the scheduling apparatus calculates the QoS of the execution path based on the QoS of each execution device.
  • the QoS of the execution path is calculated in a plurality of implementations. This is not specifically limited herein. The following illustrates several manners of calculating the QoS of the execution path.
  • the QoS i is QoS of the i th execution device on the execution path
  • ai is a weight of the i th execution device.
  • the path b 1 of the service B in FIG. 2 is still used as an example for description.
  • QoS of the VPC that is calculated according to a corresponding calculation formula 11 is QoS 1 and a weight of the VPC is 0.1
  • QoS of the ToR that is calculated according to a corresponding calculation formula 12 is QoS 2 and a weight of the ToR is 0.1
  • QoS of the bare metal host 1 that is calculated according to a corresponding calculation formula 13 is QoS 3 and a weight of the bare metal host 1 is 0.5
  • QoS of the SAN that is calculated according to a corresponding calculation formula 14 is QoS 4 and a weight of the SAN is 0.3.
  • the user may independently set QoS of a service (which may also be referred to as a to-be-deployed service).
  • the scheduling apparatus may first intelligently recommend generally used QoS (which may be referred to as general QoS) for the to-be-deployed service to the user based on a service type and a property of the to-be-deployed service that is entered by the user. Then the user sets the QoS of the to-be-deployed service by referring to the recommended general QoS (the user may directly use the general QoS as the preset QoS without modification, or may use the general QoS as the preset QoS after slightly adjusting the general QoS based on a preference of the user.
  • general QoS which may be referred to as general QoS
  • the user may directly set the QoS of the to-be-deployed service based on a requirement of the user, to obtain the preset QoS.
  • a basis for setting the preset QoS is not limited herein.
  • the scheduling apparatus may obtain the preset QoS of the to-be-deployed service.
  • step 304 may be performed before step 301, may be performed after step 303, may be performed between any two steps of step 301 to step 303, or may be performed simultaneously with any one of step 301 to step 303. This is not specifically limited herein.
  • the scheduling apparatus After obtaining the QoS of the to-be-deployed service on the execution path and the QoS preset by the user for the to-be-deployed service, the scheduling apparatus determines whether the relationship between the QoS of the execution path and the preset QoS meets the preset rule. If the scheduling apparatus determines that the relationship between the QoS of the execution path and the preset QoS meets the preset rule, the scheduling apparatus deploys the to-be-deployed service on the execution path.
  • the preset rule may be that the QoS of the execution path is greater than the preset QoS, or a value obtained by subtracting the preset QoS from the QoS of the execution path is greater than a preset positive value.
  • the preset rule is not specifically limited herein.
  • This embodiment of this application first describes how to obtain the QoS of the path (namely, the execution path) on which the to-be-deployed service can be executed. Whether the to-be-deployed service is finally deployed on the execution path during execution is determined by comparing the preset QoS of the to-be-deployed service with the QoS of the execution path. In addition, this embodiment of this application further provides for the first time a solution in which the QoS of the to-be-deployed service may be set by the user. This fully considers preferences of different users for QoS of a same to-be-deployed service, and improves service satisfaction of the to-be-deployed service.
  • the scheduling apparatus may further obtain the QoS of the transmission link between the execution devices on the execution path before step 303.
  • the QoS of the execution path obtained in step 303 is calculated based on the QoS of the execution devices on the execution path and the QoS of the transmission links on the execution path.
  • the scheduling apparatus may obtain the QoS of the execution devices on the execution path in a manner described in the embodiment corresponding to FIG. 3 . Details are not described herein again.
  • how the scheduling apparatus obtains the QoS of the transmission links is specifically described as follows: First, the scheduling apparatus determines the transmission links on the execution path, determines a QoS calculation formula corresponding to each transmission link, and then sends the QoS calculation formula corresponding to each transmission link to one of two execution devices on the corresponding transmission link.
  • the execution device that obtains the QoS calculation formula calculates QoS of a corresponding transmission link, and sends the QoS of the transmission link to the scheduling apparatus.
  • the scheduling apparatus may determine that transmission links of the path b1 includes a transmission link 1, a transmission link 2, and a transmission link 3, and determines that QoS calculation formulas corresponding to the transmission link 1, the transmission link 2, and the transmission link 3 are respectively a calculation formula 21, a calculation formula 22, and a calculation formula 23, where two execution devices on the transmission link 1 are the VPC and the ToR, two execution devices on the transmission link 2 are the ToR and the bare metal host 1, and two execution devices on the transmission link 3 are the bare metal host 1 and the SAN.
  • the scheduling apparatus may send the calculation formula 21 to the VPC or the ToR (it is assumed that the calculation formula is sent to the VPC), send the calculation formula 22 to the ToR or the bare metal host 1 (it is assumed that the calculation formula is sent to the bare metal host 1), and send the calculation formula 23 to the bare metal host 1 or the SAN (it is assumed that the calculation formula is sent to the SAN).
  • the VPC, the bare metal host 1, and the SAN respectively calculate QoS of the transmission link 1, the transmission link 2, and the transmission link 3 according to the calculation formula 21, the calculation formula 22, and the calculation formula 23.
  • the VPC, the bare metal host 1, and the SAN respectively send the obtained QoS of the three transmission links to the scheduling apparatus.
  • the scheduling apparatus may further send a QoS calculation formula corresponding to a transmission link to two execution devices on the transmission link, so that the two execution devices both calculate QoS of the same transmission link.
  • the scheduling apparatus takes an arithmetic average value of the obtained two QoSs of the same transmission link as QoS of the transmission link on the execution path.
  • the scheduling apparatus in addition to the QoS of the execution devices on the execution path, the scheduling apparatus further obtains the QoS of the transmission link between the execution devices, and determines the QoS of the execution path based on the QoS of the execution devices and the QoS of the transmission link, so that the calculated QoS of the entire execution path is more accurate.
  • the scheduling apparatus may determine the execution path according to but is not limited to the following manner: First, the scheduling apparatus determines N (N ⁇ 1) execution paths based on the service identifier, where specifications of resources included in execution devices on each execution path are not lower than specifications of resources required by the to-be-deployed service. For a method for determining QoS of each of the N execution paths by the scheduling apparatus and a method for obtaining the QoS of each execution path by the scheduling apparatus, refer to steps 301 to 305 in the foregoing embodiment. Details are not described herein.
  • the scheduling apparatus first determines several execution paths (for example, a path d1, a path d2, and a path d3 can provide, for a service D, bandwidth resources with specifications not lower than 100 kbps) based on the specifications (for example, specifications of bandwidth resources required by the service D are not lower than 100 kbps) of the resources required by the to-be-deployed service.
  • a path d1, a path d2, and a path d3 can provide, for a service D, bandwidth resources with specifications not lower than 100 kbps) based on the specifications (for example, specifications of bandwidth resources required by the service D are not lower than 100 kbps) of the resources required by the to-be-deployed service.
  • the service B corresponding to FIG. 2 is still used as an example for description. If the scheduling apparatus determines that execution paths that meet a resource value required by the service B include not only the path b1, but also a path b2 and a path b3, the execution path of the service B is any one of the three execution paths.
  • the scheduling apparatus may separately obtain QoS of the path b1, the path b2, and the path b3 according to the manner of obtaining the QoS of the execution path in the foregoing embodiment, then compare the obtained QoS of each of the path b1, the path b2, and the path b3 with preset QoS preset by the user for the service B, and determine, based on a comparison result, an execution path on which the service B is finally deployed. The following is described based on the comparison result.
  • the service B corresponding to FIG. 2 is still used as an example for description. If a relationship between the QoS of only one (for example, the path b1) of the path b1, the path b2, and the path b3 and the preset QoS meets the preset rule, the scheduling apparatus deploys the service B on the path b1.
  • the scheduling apparatus may further perform prediction on the QoS (the QoS of the execution path mentioned above is QoS of the execution path in a current periodicity) of the execution path to obtain predicted QoS of the execution path in a next periodicity. After determining that a relationship between the predicted QoS of the execution path and the preset QoS of the to-be-deployed service also meets the preset rule, the scheduling apparatus deploys the to-be-deployed service on the execution path.
  • the service B is used as an example.
  • the scheduling apparatus may further perform prediction on the QoS of the path b1 to obtain predicted QoS of the path b1 in a next periodicity. After determining that a relationship between the predicted QoS of the path b1 and the preset QoS of the service B also meets the preset rule, the scheduling apparatus deploys the service B on the path b1. b. If the scheduling apparatus determines that a relationship between QoS of at least two of the N execution paths and the preset QoS meets the preset rule, the scheduling apparatus deploys the service on one of the at least two execution paths.
  • the service B corresponding to FIG. 2 is still used as an example for description. If a relationship between QoS of some (for example, the path b1 and the path b2) or all of the path b1, the path b2, and the path b3 and the preset QoS meets the preset rule (the execution paths that meet the preset rule may be collectively referred to as to-be-selected execution paths), a to-be-deployed execution path of the service B may be determined in the following manners:
  • the scheduling apparatus may adjust the resources of the N execution paths based on the resource value required by the to-be-deployed service, adjust specifications of resources in execution devices on any execution path until an execution path (which may be referred to as an adjusted execution path) that meets a preset QoS condition is obtained, and then deploy the to-be-deployed service on the adjusted execution path.
  • the service B corresponding to FIG. 2 is still used as an example for description. If the scheduling apparatus finds that a relationship between QoS of the service B that is calculated on each optional execution path (for example, the path b1, the path b2, and the path b3) and the preset QoS does not meet the preset rule, backend resource adjustment is triggered.
  • the backend of the scheduling apparatus may reduce, by migration, suspension, limitation, or the like, usage of resources occupied by each execution device on the execution path of another service (for example, the service D and the service E) that is deployed on the optional execution path of the service B, to reduce contention of resources of the service B, and upgrades the QoS of the service B on the execution devices, so that the QoS of the service B on an execution path formed by the execution devices is also upgraded. That is, the QoS of the service B on the execution path may be upgraded by changing resource reservation or reducing resource contention of another service on the execution devices on the execution path.
  • another service for example, the service D and the service E
  • a manner in which the scheduling apparatus performs resource adjustment may include:
  • the scheduling apparatus increases the specifications of the resources included in the execution devices on the execution path by migration, suspension, limitation, or the like to upgrade the QoS of the execution devices.
  • the scheduling apparatus may further add additional resources to the execution devices on the execution path, to upgrade the QoS of the execution devices.
  • a resource adjustment manner is not specifically limited herein. Then, after the resource adjustment, QoS of each optional execution path (namely, the path b 1, the path b2, and the path b3) of the service B may be changed, the changed QoS of each execution path is compared with the preset QoS.
  • the scheduling apparatus may determine the to-be-deployed execution path of the service B in a similar manner in the foregoing embodiment. Alternatively, if the scheduling apparatus determines that a relationship between QoS of any execution path after the resource adjustment and the preset QoS still does not meet the preset rule, the scheduling apparatus continues to perform the resource adjustment until a relationship between QoS of at least one execution path after the resource adjustment and the preset QoS meets the preset rule.
  • FIG. 4 is used as an example for description.
  • the scheduling apparatus determines that a relationship between QoS of the path b2 of the service B and the preset QoS meets the preset rule, and then deploys the service B on the path b2.
  • the path b2 includes a VPC, a high-quality multi-layer open virtual switch OVS, a virtual machine 2, distributed block storage, a transmission link 4 connecting a VTEP 3 of the OVS and a VTEP 1 of the VPC, a transmission link 5 connecting the OVS and the virtual machine 2, and a transmission link 6 connecting the virtual machine 2 and the distributed block storage.
  • a process of determining the to-be-deployed execution path of the to-be-deployed service in the foregoing embodiment is as follows: The scheduling apparatus first determines, based on the identifier of the to-be-deployed service, the N execution paths that meet the specifications of the resources required by the to-be-deployed service, calculates N QoSs of the N execution paths, further determines that relationships between the preset QoS and several QoSs in the N QoSs meet the preset rule (it is assumed that W QoSs meet the preset rule), finally, selects an execution path corresponding to one of the W QoSs in the foregoing determining manner, and deploys the to-be-deployed service on the selected execution path.
  • the scheduling apparatus may calculate the QoS of the N execution paths one by one according to a preset method. Each time the scheduling apparatus calculates QoS of one execution path, the scheduling apparatus compares the QoS of the execution path with the preset QoS. When determining that a relationship between the QoS of the execution path and the preset QoS meets the preset rule, the scheduling apparatus directly deploys the to-be-deployed service on the execution path that meets the requirement, and no longer calculates QoS of remaining execution paths. This reduces a calculation amount.
  • the service B corresponding to FIG. 2 is used as an example for description.
  • the scheduling apparatus continues to calculate the QoS of the path b2, and then compares the QoS of the path b2 with the preset QoS. Assuming that a comparison result is that a relationship between the QoS of the path b2 and the preset QoS meets the preset rule, the scheduling apparatus directly determines that the path b2 is a to-be-deployed path of the service B, deploys the service B on the path b2, and does not calculate the QoS of the path b3.
  • the scheduling apparatus may further directly determine an execution path based on the specifications of the resources required by the to-be-deployed service (that is, the scheduling apparatus does not need to determine the N execution paths that meet the specifications of the resources of the to-be-deployed first). Each time determining an execution path, the scheduling apparatus calculates QoS of the execution path, and compares the QoS of the execution path with the preset QoS.
  • the scheduling apparatus determines that a relationship between the QoS of the execution path and the preset QoS meets the preset rule, the scheduling apparatus directly determines that the execution path is the to-be-deployed execution path of the to-be-deployed service, deploys the to-be-deployed service on this execution path, and no longer determines whether there is another execution path meets the specifications of the resources required by the to-be-deployed service.
  • the service B corresponding to FIG. 2 is still used as an example for description.
  • the scheduling apparatus first determines an execution path (it is assumed that the execution path is the path b1) of the current service B based on specifications of resources required by the service B, the scheduling apparatus directly calculates the QoS of the path b1, and compares the QoS of the path b1 with the preset QoS. If a comparison result is that a relationship between the QoS of the path b1 and the preset QoS does not meet the preset rule, the scheduling apparatus continues to determine another execution path (it is assumed that the execution path is the path b2) of the service B based on the specifications of the resources required by the service B, and calculates the QoS of the path b2.
  • the scheduling apparatus directly determines that the path b2 is the to-be-deployed execution path of the service B, deploys the service B on the path b2, and no longer determines whether there is another execution path can meet the specifications of the resources required by the service B.
  • the scheduling apparatus determines that a relationship between QoS of each of all optional execution paths (namely, the N execution paths) of the to-be-deployed service and the preset QoS does not meet the preset rule, the scheduling apparatus performs the resource adjustment to implement scheduling across execution paths such as a virtual machine, a container, and a physical machine.
  • a distributed resource scheduler distributed resource scheduler, DRS
  • DRS distributed resource scheduler
  • the DRS system can only implement virtual machine migration based on host performance, but cannot implement centralized scheduling across the virtual machine, the container, and the physical machine.
  • a virtual machine VMware is used as an example.
  • a DRS of the VMware can perform backend resource adjustment only based on a basic CPU and memory, but cannot switch and adjust an execution path of a to-be-deployed service.
  • the scheduling apparatus may classify execution paths of all services that can be executed on the cloud server in advance. For example, the scheduling apparatus classifies execution paths that can provide high storage resources into a category, classifies execution paths that can provide high network resources into a category, classifies execution paths that can provide high compute resources into a category, and the like.
  • the execution paths in each category may be classified into several subcategories based on a range of resources that can be provided. Each category and each subcategory have a corresponding identifier.
  • the scheduling apparatus may more quickly determine, based on category identifiers, all optional execution paths (namely, N execution paths) that meet the specifications of the resources required by the to-be-deployed service. This improves a speed at which the scheduling apparatus determines an execution path for deploying the to-be-deployed service.
  • the scheduling apparatus may adjust resources of execution devices on an execution path, to upgrade QoS that is calculated by the execution devices according to QoS calculation formulas, and the scheduling apparatus may further adjust the QoS calculation formulas corresponding to the execution devices, to upgrade QoS of the execution devices.
  • a premise of adjusting the QoS calculation formulas by the scheduling apparatus is that specifications of resources included in the execution path for the to-be-deployed service are actually sufficient, but the resources included in the execution devices are not fully utilized only due to different selections on which the QoS calculation formulas are based.
  • QoS of a host of the execution device may be calculated according to a QoS calculation formula of central processing unit (CPU) usage, memory usage, a bandwidth, a latency, and the like of the host.
  • the QoS calculation formulas vary depending on selections such as the CPU usage, the memory usage, the bandwidth, and the latency.
  • the execution device may obtain, through an agent (which is also referred to as Agent, and is a software/hardware entity) embedded in the execution device, a QoS calculation formula corresponding to the execution device and/or a QoS calculation formula corresponding to a transmission link that are/is sent by the scheduling apparatus, and send, back to the scheduling apparatus, QoS of the execution device and/or QoS of the transmission link that are/is calculated according to the QoS calculation formulas.
  • Agent which is also referred to as Agent, and is a software/hardware entity
  • FIG. 5 is used as an example for description. If the scheduling apparatus determines P execution paths based on a resource value required by a service E, where P ⁇ 1, the P execution paths include n execution devices.
  • Agents are embedded in the n execution devices, and may be denoted as an agent-1, an agent-2, ..., and an agent-n.
  • the scheduling apparatus After determining the n execution devices on the P execution paths, the scheduling apparatus sends, to the agents of the corresponding execution devices, QoS calculation formulas (including a QoS calculation formula corresponding to each execution device and/or a QoS calculation formula corresponding to each transmission link) corresponding to the execution devices. That is, steps p1 to p3 in FIG.
  • p1 Send, to the agent-1 on an execution device 1, QoS calculation formulas 1 (including a QoS calculation formula 1 of the execution device 1 and/or a QoS calculation formula 1 of a transmission link 1 corresponding to the execution device 1) corresponding to the execution device 1;
  • p2 Send, to the agent-2 on an execution device 2, QoS calculation formulas 2 (including a QoS calculation formula 2 of the execution device 2 and/or a QoS calculation formula 2 of a transmission link 2 corresponding to the execution device 2) corresponding to the execution device 2; ...;
  • p3 Send, to the agent-n on an execution device n, QoS calculation formulas n (including a QoS calculation formula n of the execution device n and/or a QoS calculation formula n of a transmission link n corresponding to the execution device n) corresponding to the execution device n.
  • the agent on each execution device After obtaining the corresponding QoS calculation formula, the agent on each execution device obtains QoS of the corresponding execution device according to the QoS calculation formula (if the QoS calculation formula includes only the QoS calculation formula corresponding to the execution device) and QoS of the transmission link corresponding to the execution device (if the QoS calculation formula includes the QoS calculation formula corresponding to the execution device and the QoS calculation formula corresponding to the transmission link). That is, steps p4 to p6 in FIG.
  • each device sends the calculated QoS of the execution device and the calculated QoS of the transmission link corresponding to the execution device to the scheduling apparatus.
  • steps p7 to p9 in FIG. 5 are performed: p7: The agent-1 sends the QoS of the execution device 1 and the QoS of the transmission link 1 to the scheduling apparatus; p8: The agent-2 sends the QoS of the execution device 2 and the QoS of the transmission link 2 to the scheduling apparatus; ... ; p9: The agent-n sends the QoS of the execution device n and the QoS of the transmission link n to the scheduling apparatus.
  • the scheduling apparatus may calculate QoS of the service E on the P execution paths based on the QoS of the execution devices and the QoS of the transmission links. It should be noted that, in FIG.
  • step p7 is performed after step p4
  • step p4 is performed after step p1
  • step p8 is performed after step p5
  • step p9 is performed after step p6
  • step p6 is performed after step p3. This is not specifically limited herein.
  • the scheduling apparatus sends and receives the QoS calculation formulas, and calculates and sends QoS through the agents embedded in the execution devices. This facilitates centralized management. If a new execution device needs to be added to the cloud server, centralized management can be implemented only by embedding an agent in the newly added execution device, and this is simple and convenient.
  • the portal is a portal web, and is a web-based application that mainly provides personalization, single sign-on, content integration of different sources, and a presentation layer for storing an information system), and enters the to-be-deployed service.
  • a scheduling apparatus obtains related features of the to-be-deployed service.
  • the scheduling apparatus may obtain the related features such as a type, a property, and a priority of the to-be-deployed service.
  • the scheduling apparatus intelligently recommends general QoS of the to-be-deployed service to the user.
  • the scheduling apparatus intelligently recommends the general QoS of the to-be-deployed service, namely, the QoS commonly used when most users execute the to-be-deployed service, to the user based on the related features of the to-be-deployed service.
  • the user sets preset QoS of the to-be-deployed service.
  • the user sets the QoS of the to-be-deployed service with reference to the general QoS recommended by the scheduling apparatus.
  • the user may directly use the general QoS as the preset QoS without modification, or may use the general QoS as the preset QoS after slightly adjusting the general QoS based on a preference of the user. This is not specifically limited herein.
  • step 603 may be skipped, and the user may directly set the QoS of the to-be-deployed service based on a requirement of the user, to obtain the preset QoS.
  • the scheduling apparatus obtains the preset QoS set by the user.
  • the scheduling apparatus may obtain the preset QoS of the to-be-deployed service.
  • the scheduling apparatus determines specifications of resources required by the to-be-deployed service.
  • the scheduling apparatus may further determine the specifications of the resources required by the to-be-deployed service based on the obtained related features of the to-be-deployed service. It should be noted that step 606 may be performed after step 602 to step 605, and step 606 may also be performed between any two steps in step 602 to step 605. This is not specifically limited herein.
  • the scheduling apparatus determines Q execution paths of the to-be-deployed service and m QoS calculation formulas corresponding to m execution devices on the Q execution paths.
  • the scheduling apparatus After determining a resource value required by the to-be-deployed service, the scheduling apparatus automatically generates, based on an overall system status of a cloud server, the Q execution paths (Q ⁇ 1) that meet the specifications of the resources required by the to-be-deployed service, and further determines the m execution devices included in the Q execution paths and the m QoS calculation formulas corresponding to the m execution devices, where the m QoS calculation formulas include m QoS calculation formulas corresponding to the m execution devices and QoS calculation formulas corresponding to m transmission links.
  • the QoS calculation formulas corresponding to the execution devices are used to calculate QoS of the corresponding execution devices, and the QoS calculation formulas corresponding to the transmission links are used to calculate QoS of the transmission links corresponding to the execution devices.
  • definitions of the QoS calculation formula corresponding to the execution device and the QoS calculation formula corresponding to the transmission link are described in the embodiment corresponding to FIG. 3 . Details are not described herein again.
  • the scheduling apparatus determines weights of execution devices on each of the Q execution paths.
  • the scheduling apparatus After determining the execution devices on each of the Q execution paths, the scheduling apparatus further determines the weight of each execution device, where the weight of the execution device indicates an importance degree of the execution device on a corresponding execution path.
  • the scheduling apparatus sends the m corresponding QoS calculation formulas to agents embedded in the m execution devices.
  • an agent is embedded in each execution device on the cloud server, and the execution device sends and receives the QoS calculation formula, and calculates and sends the QoS through the embedded agent.
  • the scheduling apparatus determines the m execution devices of the Q execution paths, the scheduling apparatus sends, to the agents of the m execution devices, the QoS calculation formulas corresponding to the m execution devices.
  • Each agent on the m execution devices calculates, according to corresponding QoS calculation formulas, QoS of a corresponding execution device and QoS of a transmission link corresponding to the execution device.
  • each agent on the m execution devices calculates, according to the QoS calculation formulas, the QoS of the corresponding execution device and the QoS of the transmission link corresponding to the execution device.
  • how the agent on the execution device specifically obtains, according to the QoS calculation formulas, the QoS of the corresponding execution device and the QoS of the transmission link corresponding to the execution device is similar to a manner of calculating the QoS of the execution device and the QoS of the transmission link in the embodiment corresponding to FIG. 3 . Details are not described herein again.
  • Each agent on the m execution devices sends the QoS of the corresponding execution device and the QoS of the corresponding transmission link to the scheduling apparatus.
  • Each agent on the m execution devices sends, to the scheduling apparatus, the QoS of the corresponding execution device and the QoS of the corresponding transmission link that are obtained according to the QoS calculation formulas.
  • the scheduling apparatus calculates QoS of the Q execution paths, and predicts QoS of each execution path in a next periodicity.
  • the scheduling apparatus calculates the QoS of the Q execution paths based on the obtained QoS of the m execution devices and the obtained QoS of the corresponding transmission links, and predicts the QoS of each execution path in the next periodicity.
  • how the scheduling apparatus calculates the QoS of the Q execution paths is similar to a manner of calculating the QoS of the execution path in the embodiment corresponding to FIG. 3 . Details are not described herein again.
  • a final execution path of the to-be-deployed service may be determined according to the solution described in the foregoing embodiment. Details are not described herein.
  • the scheduling apparatus determines an execution path as a to-be-deployed execution path of the to-be-deployed service based on QoS of the execution path in a current periodicity, execution devices on the execution path not only run the to-be-deployed service, but may also run another service. Therefore, resources included in the execution devices may change.
  • the scheduling apparatus finds, in a process of executing the to-be-deployed service, that the selected execution path cannot meet a resource requirement of the to-be-deployed service, the scheduling apparatus needs to perform a non-prediction-based backend adjustment on the resources on the execution path. For example, when data of a service F is transmitted to an execution path such as a unified load balancer, the scheduling apparatus finds that virtual machine resources at the backend of the execution path are insufficient, so that a requirement for latency calculation of the service F cannot be met. In this case, the unified load balancer automatically forwards the service F to a physical machine for calculation. The scheduling apparatus may further predict QoS of each execution path in the next periodicity.
  • the scheduling apparatus may adjust the execution path of the to-be-deployed service based on the predicted QoS of each execution path in the next periodicity to an execution path whose QoS meets the preset rule in the next periodicity, so as to ensure continuity and stability of the to-be-deployed service.
  • the scheduling apparatus performs resource adjustment.
  • the scheduling apparatus determines that the relationship between the calculated QoS of each of the Q execution paths and the preset QoS does not meet the preset rule, the scheduling apparatus performs the resource adjustment, that is, adjust specifications of resources of execution devices on any execution path until an execution path (which may be referred to as the adjusted execution path) that meets a preset QoS condition is obtained, and deploys the to-be-deployed service on the adjusted execution path.
  • the resource adjustment that is, adjust specifications of resources of execution devices on any execution path until an execution path (which may be referred to as the adjusted execution path) that meets a preset QoS condition is obtained, and deploys the to-be-deployed service on the adjusted execution path.
  • a manner in which the scheduling apparatus performs the resource adjustment is similar to the manner in which the scheduling apparatus performs the resource adjustment in the foregoing embodiment. Details are not described herein again.
  • execution paths may be different due to different compute resources.
  • the scheduling apparatus may be divided into functional modules based on the foregoing examples of the service deployment method.
  • each functional module may be obtained through division based on each corresponding function, or two or more functions may be integrated into one processing module.
  • the integrated module may be implemented in a form of hardware, or may be implemented in a form of a software functional module.
  • module division is an example, and is merely a logical function division. In actual implementation, another division manner may be used.
  • FIG. 7 is a schematic diagram of a scheduling apparatus.
  • the scheduling apparatus provided in an embodiment of this application may include: a path selection module 701, configured to:
  • the path selection module 701 is specifically configured to: before calculating the QoS of the execution path based on the QoS of each execution device, obtain QoS of at least one transmission link between the at least two execution devices, and calculate the QoS of the execution path based on the QoS of each execution device and the QoS of the at least one transmission link.
  • the path selection module 701 is further specifically configured to: determine at least two first QoS calculation formulas corresponding to each execution device, where one execution device corresponds to one QoS calculation formula, send each first QoS calculation formula to a corresponding execution device, and finally obtain the QoS of each execution device that is obtained by each execution device according to a corresponding first QoS calculation formula.
  • the path selection module 701 is specifically configured to: determine at least one second QoS calculation formula corresponding to the at least one transmission link, send each second QoS calculation formula to an execution device on a corresponding transmission link, and finally obtain QoS of each transmission link that is obtained by the execution device on the at least one transmission link according to a corresponding second QoS calculation formula.
  • the path selection module 701 is specifically configured to: determine a weight of the QoS of each execution device, and determine the QoS of the execution path based on the QoS of each execution device and the corresponding weight.
  • the path selection module 701 is specifically configured to: determine N execution paths based on resources required by the to-be-deployed service, where specifications of resources included in execution devices on each execution path are not lower than specifications of the resources required by the to-be-deployed service; if a relationship between QoS of each of the N execution paths and the preset QoS does not meet the preset rule, perform resource adjustment, that is, adjust specifications of resources in execution devices on any execution path until an execution path that meets a preset QoS condition is obtained.
  • the path selection module 701 is further specifically configured to: predict QoS of the execution path in a next periodicity, to obtain predicted QoS of the execution path, where the QoS of the execution path is the QoS of the execution path in a current periodicity; and determine that a relationship between the predicted QoS of the execution path and the preset QoS also meets the preset rule.
  • the service deployment module 702 is specifically configured to: after the path selection module determines that the relationship between the predicted QoS of the execution path and the preset QoS meets the preset rule, deploy the to-be-deployed service on the execution path.
  • a specific function and a structure of the scheduling apparatus in the embodiment corresponding to FIG. 7 are used to implement the steps performed by the scheduling apparatus in FIG. 3 to FIG. 6A and FIG. 6B . Details are not described herein again.
  • FIG. 7 describes the scheduling apparatus in the embodiments of this application from a perspective of a modular functional entity.
  • the following describes the scheduling apparatus in the embodiments of this application from a perspective of hardware processing.
  • FIG. 8 is a schematic diagram of an embodiment of a scheduling apparatus according to an embodiment of this application.
  • the scheduling apparatus includes the following features.
  • the scheduling apparatus may vary greatly with different configurations or performance, and may include one or more central processing units (CPU) 822 (for example, one or more processors) and one or more memories 830, where one or more computer instructions 842 or data 844 is stored on the memory 830.
  • the memory 830 may perform transient storage or persistent storage.
  • the memory 830 may include one or more modules (not shown in the figure), and each module may include an operation on a series of computer instructions in the scheduling apparatus.
  • the central processing unit 822 may be configured to communicate with the memory 830, and execute, on the scheduling apparatus, the operations of the series of computer instructions in the memory 830.
  • the scheduling apparatus may further include one or more power supplies 826, one or more wired or wireless network interfaces 850, one or more input/output interfaces 858, and/or one or more operating systems 841.
  • Steps in the service deployment methods described in FIG. 3 to FIG. 6A and FIG. 6B are implemented by the scheduling apparatus based on the structure shown in FIG. 8 .
  • All or some of the foregoing embodiments may be implemented by using software, hardware, firmware, or any combination thereof.
  • the embodiments may be implemented completely or partially in a form of a computer program product.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable apparatuses.
  • the computer instructions may be stored in a computer-readable storage medium or may be transmitted from a computer-readable storage medium to another computer-readable storage medium.
  • the computer instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired (for example, a coaxial cable, an optical fiber, or a digital subscriber line) or wireless (for example, infrared, radio, or microwave) manner.
  • the computer-readable storage medium may be any usable medium accessible by a computer, or a data storage device, such as a server or a data center, integrating one or more usable media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, or a magnetic tape), an optical medium (for example, a DVD), a semiconductor medium (for example, a solid-state drive), or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
EP20834405.1A 2019-06-29 2020-06-04 Method for deploying services and scheduling apparatus Active EP3968606B1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910581292.5A CN110430236B (zh) 2019-06-29 2019-06-29 一种部署业务的方法以及调度装置
PCT/CN2020/094442 WO2021000694A1 (zh) 2019-06-29 2020-06-04 一种部署业务的方法以及调度装置

Publications (4)

Publication Number Publication Date
EP3968606A1 EP3968606A1 (en) 2022-03-16
EP3968606A4 EP3968606A4 (en) 2022-06-22
EP3968606B1 true EP3968606B1 (en) 2024-03-20
EP3968606C0 EP3968606C0 (en) 2024-03-20

Family

ID=68408881

Family Applications (1)

Application Number Title Priority Date Filing Date
EP20834405.1A Active EP3968606B1 (en) 2019-06-29 2020-06-04 Method for deploying services and scheduling apparatus

Country Status (4)

Country Link
US (1) US11838389B2 (zh)
EP (1) EP3968606B1 (zh)
CN (1) CN110430236B (zh)
WO (1) WO2021000694A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110430236B (zh) * 2019-06-29 2020-11-03 华为技术有限公司 一种部署业务的方法以及调度装置
CN114172853B (zh) * 2021-12-01 2024-04-16 阿里巴巴（中国）有限公司 流量转发及裸机服务器的配置方法、装置

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1892882A1 (en) * 2006-08-25 2008-02-27 Alcatel Lucent Method of providing QoS for a flow
US8982709B2 (en) * 2007-01-31 2015-03-17 Hewlett-Packard Development Company, L.P. Selecting service nodes for an end-to-end service path from a reduced search space
KR101054055B1 (ko) * 2008-05-16 2011-08-03 한국과학기술원 이동 노드를 위한 ｎｓｉｓ 프로토콜 기반 사전자원예약장치 및 방법
CN102204209B (zh) * 2011-05-06 2014-01-22 华为技术有限公司 跨媒介网络中的QoS协商方法和系统及网络设备
CN102404137A (zh) * 2011-08-09 2012-04-04 江苏欣网视讯科技有限公司 一种IP网络下基于业务流QoS监测的故障定位方法
WO2015168888A1 (zh) * 2014-05-07 2015-11-12 华为技术有限公司 网络拥塞控制方法和控制器
CN104104602B (zh) * 2014-08-06 2018-02-09 湖南网数科技有限公司 一种QoS信息预测的方法及装置
CN104518993A (zh) * 2014-12-29 2015-04-15 华为技术有限公司 云化网络通信路径的分配方法、装置及系统
CN104539535A (zh) * 2015-01-28 2015-04-22 浪潮电子信息产业股份有限公司 一种数据传输路径确定方法及装置
CN108243044B (zh) * 2016-12-27 2021-08-24 山东财兜信息科技有限公司 业务部署的方法与装置
CN108494601B (zh) * 2018-04-04 2020-02-21 西安电子科技大学 层次化确定网络中的多约束双路径路由方法
CN109729557A (zh) * 2018-12-28 2019-05-07 上海昭宁信息科技有限公司 一种无线组网的方法和终端
US10979316B2 (en) * 2019-05-31 2021-04-13 Juniper Networks, Inc. Dynamic application SLA metric generation, distribution, and intent-based SD-WAN link selection
CN110430236B (zh) * 2019-06-29 2020-11-03 华为技术有限公司 一种部署业务的方法以及调度装置

Also Published As

Publication number Publication date
EP3968606A1 (en) 2022-03-16
CN110430236A (zh) 2019-11-08
US20220124172A1 (en) 2022-04-21
WO2021000694A1 (zh) 2021-01-07
EP3968606A4 (en) 2022-06-22
US11838389B2 (en) 2023-12-05
CN110430236B (zh) 2020-11-03
EP3968606C0 (en) 2024-03-20

Similar Documents

Publication Publication Date Title
CN112970228B (zh) 提供网络切片服务时具有冲突管理的性能保证方法和系统
JP6823203B2 (ja) ネットワークスライスを作成するための方法及び装置並びに通信システム
EP1472846B1 (en) Method and apparatus for web farm traffic control
US9417903B2 (en) Storage management for a cluster of integrated computing systems comprising integrated resource infrastructure using storage resource agents and synchronized inter-system storage priority map
US7680897B1 (en) Methods and systems for managing network traffic
CN110463140B (zh) 计算机数据中心的网络服务水平协议
US11838389B2 (en) Service deployment method and scheduling apparatus
US7627618B2 (en) System for managing data collection processes
Ascigil et al. Resource provisioning and allocation in function-as-a-service edge-clouds
US20120079147A1 (en) Bus control device
CN113348651B (zh) 切片的虚拟网络功能的动态云间放置
US10609118B2 (en) Adaptive communication control device
US20140089510A1 (en) Joint allocation of cloud and network resources in a distributed cloud system
CN112119666A (zh) 用于管理无线电接入网内的资源的方法、计算机程序和电路系统
CN112822050A (zh) 用于部署网络切片的方法和装置
US20140189092A1 (en) System and Method for Intelligent Data Center Positioning Mechanism in Cloud Computing
CN110086726A (zh) 一种自动切换Kubernetes主节点的方法
US20220329651A1 (en) Apparatus for container orchestration in geographically distributed multi-cloud environment and method using the same
EP3698246A1 (en) Management of a virtual network function
Ben Jemaa et al. Analytical models for QoS-driven VNF placement and provisioning in wireless carrier cloud
WO2020108337A1 (zh) 一种cpu资源调度方法及电子设备
EP3855689A1 (en) Method, apparatus, and system for providing service, storage medium, and electronic device
Casetti et al. Arbitration among vertical services
Schneider et al. Trade-offs in dynamic resource allocation in network function virtualization
Taka et al. Service placement and user assignment in multi-access edge computing with base-station failure

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20211207

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: HUAWEI CLOUD COMPUTING TECHNOLOGIES CO., LTD.

A4 Supplementary search report drawn up and despatched

Effective date: 20220524

RIC1 Information provided on ipc code assigned before grant

Ipc: H04L 67/50 20220101ALI20220518BHEP

Ipc: H04L 41/5054 20220101AFI20220518BHEP

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
REG Reference to a national code

Ref document number: 602020027655

Country of ref document: DE

Ref country code: DE

Ref legal event code: R079

Free format text: PREVIOUS MAIN CLASS: H04L0029080000

Ipc: H04L0041505400

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

RIC1 Information provided on ipc code assigned before grant

Ipc: H04L 67/61 20220101ALI20231026BHEP

Ipc: H04L 67/10 20220101ALI20231026BHEP

Ipc: H04L 67/50 20220101ALI20231026BHEP

Ipc: H04L 41/5054 20220101AFI20231026BHEP

INTG Intention to grant announced

Effective date: 20231115

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602020027655

Country of ref document: DE

U01 Request for unitary effect filed

Effective date: 20240320

U07 Unitary effect registered

Designated state(s): AT BE BG DE DK EE FI FR IT LT LU LV MT NL PT SE SI

Effective date: 20240322